A Comparative Analysis of Kernel-Based Methods for the Classification of Land Cover Maps in Satellite Imagery

Author

Basili, R. ; Del Frate, F. ; Luciani, M. ; Mesiano, F. ; Pacifici, F. ; Rossi, R.

Author_Institution

Dept. of Comput. Sci., Tor Vergata Univ., Rome

Volume

4

fYear

2008

fDate

7-11 July 2008

Abstract

This paper studies the impact of several learning issues in an image classification task with SVMs, such as rich feature-based representations, optimization and sensitivity to novelty in the test data sets. The employed imagery refers to the city of Rome, Italy and is acquired in different years and seasons by the European Remote Sensing Satellites ERS-1 and ERS-1/2 tandem mission. A comprehensive evaluation according to varying training conditions is reported, showing that SVMs provide robust and largely applicable tools.

Keywords

expert systems; geophysical signal processing; image classification; image representation; optimisation; sensitivity analysis; support vector machines; vegetation mapping; ERS-1 mission; ERS-1/2 mission; European Remote Sensing Satellites; Italy; Rome; SVM; image classification; kernel based method; land cover maps; optimization; rich feature based representation; satellite imagery; sensitivity; support vector machine; Cities and towns; Image analysis; Image classification; Kernel; Laboratories; Remote sensing; Robustness; Satellites; Support vector machine classification; Support vector machines;

fLanguage

English

Publisher

ieee

Conference_Titel

Geoscience and Remote Sensing Symposium, 2008. IGARSS 2008. IEEE International

Conference_Location

Boston, MA

Print_ISBN

978-1-4244-2807-6

Electronic_ISBN

978-1-4244-2808-3

Type

conf

DOI

10.1109/IGARSS.2008.4779827

Filename

4779827